Mu Hu
Mu Hu
Thanks for your interest! The point clouds have been projected into an image plane so they could be regarded as a depth image. All (not only "effective" points) pixels of...
> > Thanks for your interest! The point clouds have been projected into an image plane so they could be regarded as a depth image. All (not only "effective" points)...
Thanks for your interest. We previously did not concat them because (1) It was a mistake which should have been corrected. (2) The concatenation led to higher computational cost which...
HI! Of course you could make an attempt. In our two-branch (actually hourglass-styled) backbone, the second (DD) branch could be regarded as a reprocessing procedure of the first (CD) branch...
Hi! Of course you could add a [tensorboard writer](https://pytorch.org/docs/stable/tensorboard.html) to record variations during training. As for myself, I observe the error metrics on the validation set after each epoch to...
The results are recorded in the generated file _'val.csv'_. Feel free to direct me if you have more questions.
Hi, (1) during the training procedure the error metrics are accumulated in an on-line scheme. For the definition of RMSE and MAE you might refer the screenshot below from NLSPN[Park...
Hi!(1)Note that the groundtruth maps is automatically generated by accumulating 11 LiDaR consequent frames with outlier removed in Sparse CNNs [Uhrig et al. 3DV2017]. So they are semi-dense and can...
I used jet [here](https://matplotlib.org/stable/tutorials/colors/colormaps.html) and the range were set to 0~100m.
Hello, you might use `torch.where` for generating a binary mask and then sum it up to count the number of valid pixels.